SparkFun Forums 

Where electronics enthusiasts find answers.

Have questions about a SparkFun product or board? This is the place to be.
By digitalclouds
#106682
Hi,

I am rather new to this - but I was wondering what a good approach would be to wire up a Non-invasive AC current sensor (http://www.seeedstudio.com/depot/noninv ... Path=84_91) using an Op Amp to get an output voltage range between 0-3.3 volts.

I know that I can use resistors, etc. to convert to a voltage (have found some nice examples), but I have understood that an Op Amp would be a 'cleaner' solution, I just am not sure where to start looking for a design ..

Could I just use something like the sparkfun board http://www.sparkfun.com/commerce/produc ... ts_id=8883 which uses the ACS712 and then pumps it through an op-amp (if I understand correctly) .. my concern here is resolution and the 5v requirement of the ACS712 (I only have 3.3v <- using a JeeNode)

anyhow, suggestions, directions where to look, etc. would be much appreciated.

Thanks & kind regards,
Jake.
By theatrus
#106686
What you have there is a "raw" CT. The output of the sensor sold there (part number SCT-013-000) is a current output sensor. This means when 100A is going through the "primary" coil, the leads will excited to deliver up to 33mA of current.

First off, you need to convert this to a voltage - luckily all this takes is a resistor. 100ohms is a great starting point. Connect it across the CT.

You will receive the exact AC waveform current as a voltage. Next in line, you should perform a DC block - a capacitor sized appropriately to pass 50/60Hz. A 47uF electrolytic works.

Next up is the voltage gain stage. Depending on your resistor, you will receive a small voltage (peak to peak) signal. The SparkFun board you have listed will not work for you directly - you just need an op-amp, not a whole measurement IC. You ideally want to use an op-amp in an inverting configuration, as shown here: http://en.wikipedia.org/wiki/Operationa ... _amplifier

The resistors need to be sized to add the appropriate gain from the op-amp. Read the Wikipedia section several times through - in fact read the whole article :). Keep the resistance of the resistors higher (10k-100k).

Now you have a signal to send to your A/D unit. This will be an AC waveform. In order to determine the actual current, you need to oversample this value (preferably 256 times per cycle) and perform an RMS (root mean square) calculation.

Here is a great app-note from Atmel on building an energy meter on an AVR:

http://atmel.com/dyn/resources/prod_doc ... oc2566.pdf

Page 36 has a schematic and page 37 has the BOM.
By digitalclouds
#106688
Wow! Thanks theatrus .. that gives me a very good start (if not more - infact .. make that 'a whole solution' :D), very much appreciated.

Kind regards,
Jake.
By digitalclouds
#106698
Ok, still need to do all the maths - but at a quick look, would I be correct in thinking that the LMV321 (or the LMV358 that SF has - I know it is dual) would be an appropriate starting point for the op-amp?

http://www.national.com/ds/LM/LMV321.pdf

It looks like it uses low power, so should be able to run using the 3.3v.

Page 15, fig 10. showing a circuit much like you described.

Thanks & kind regards,
Jake.
By theatrus
#106701
Yes, that would be a reasonable op-amp to use.

When figuring out the gain to use, keep in mind the current crest factor so you don't clip the signal. In a perfect sine wave, this is 1.41 (root 2). So for 100A RMS, the peak signal is actually 141A.
By digitalclouds
#106763
Thanks for all the info, very much appreciated.

Just out of interest - how did you derive the 47uF for the DC Block .. I have had a quick look around and am unsure about the logic behind selecting this cap.

Once again .. thanks for everything so far (got an order in for a bunch of parts @ the moment to put it all into the tangible world :? scary!!)